Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #3076] [BUG]: In query mode, return answer citations with mach rates ranging from 0% to 3%. #1974

New Issue
Closed
opened 2026-02-22 18:27:29 -05:00 by yindo · 7 comments
yindo commented 2026-02-22 18:27:29 -05:00
Owner
Copy Link
Copy Source

Originally created by @kaanvur on GitHub (Jan 31, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3076

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

I added two files.
I selected query mode in chat settings.
Document similarity threshold
medium is set in vector database settings.
But the model still returns answers to irrelevant questions.
As citations, it shows a few examples with mach rates ranging from 0% to 3%.

Are there known steps to reproduce?

Image
Image
Image
Image
Image

Originally created by @kaanvur on GitHub (Jan 31, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3076 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? I added two files. I selected query mode in chat settings. Document similarity threshold medium is set in vector database settings. But the model still returns answers to irrelevant questions. As citations, it shows a few examples with mach rates ranging from 0% to 3%. ### Are there known steps to reproduce? ![Image](https://github.com/user-attachments/assets/80978aef-10f4-4ed8-a812-0aab7adeeb0c) ![Image](https://github.com/user-attachments/assets/793b354e-899d-4826-bbe1-e341e81df626) ![Image](https://github.com/user-attachments/assets/7e488f95-c960-4a33-bc36-5ce1be40300d) ![Image](https://github.com/user-attachments/assets/76ccbf95-c80f-41cb-914f-92426814411d) ![Image](https://github.com/user-attachments/assets/e3e41e38-6d10-47bf-974c-bbd01bfd7c4d)
yindo added the possible bug label 2026-02-22 18:27:29 -05:00
yindo closed this issue 2026-02-22 18:27:30 -05:00
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jan 31, 2025):

Turn off reranking when in query mode - that is the root cause of this (and also why it is not the default mode for a workspace)

@timothycarambat commented on GitHub (Jan 31, 2025): Turn off reranking when in query mode - that is the root cause of this (and also why it is not the default mode for a workspace)
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@kaanvur commented on GitHub (Feb 1, 2025):

Turn off reranking when in query mode - that is the root cause of this (and also why it is not the default mode for a workspace)

Thanks for your answer @timothycarambat but I couldn't find any instructions on where and how to do this setting. Can you help me?

@kaanvur commented on GitHub (Feb 1, 2025): > Turn off reranking when in query mode - that is the root cause of this (and also why it is not the default mode for a workspace) Thanks for your answer @timothycarambat but I couldn't find any instructions on where and how to do this setting. Can you help me?
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 1, 2025):

Of course, in the UI we call reranking "Search Preference" since re-ranking is very ambiguous to non-technical people. If you click on a workspace's settings (gear icon) and navigate to "Vector Database" you will find a field "search preference" which you can change from "Accuracy Optimized" back to "default"

https://docs.anythingllm.com/llm-not-using-my-docs#vector-database-settings--search-preference

@timothycarambat commented on GitHub (Feb 1, 2025): Of course, in the UI we call reranking "Search Preference" since re-ranking is very ambiguous to non-technical people. If you click on a workspace's settings (gear icon) and navigate to "Vector Database" you will find a field "search preference" which you can change from "Accuracy Optimized" back to "default" https://docs.anythingllm.com/llm-not-using-my-docs#vector-database-settings--search-preference
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@kaanvur commented on GitHub (Feb 2, 2025):

Thank you for your answer @timothycarambat . When I read your explanations, it gives the opposite impression :) I made the changes. This time it brought up more relevant information. However, none of the quotes that it said were 65% relevant to my question "egg bread recipe" have anything to do with this topic. How can I improve this?

Image
Image

@kaanvur commented on GitHub (Feb 2, 2025): Thank you for your answer @timothycarambat . When I read your explanations, it gives the opposite impression :) I made the changes. This time it brought up more relevant information. However, none of the quotes that it said were 65% relevant to my question "egg bread recipe" have anything to do with this topic. How can I improve this? ![Image](https://github.com/user-attachments/assets/a5a76e86-f2b2-4fdd-8cf8-1c2facd819ec) ![Image](https://github.com/user-attachments/assets/5fd61843-5909-4e38-b83a-e45771565e53)
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@kaanvur commented on GitHub (Feb 2, 2025):

Would changing the Embedding model here provide better results?

@kaanvur commented on GitHub (Feb 2, 2025): Would changing the Embedding model here provide better results?
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 3, 2025):

@kaanvur The default embedder is pretty bad at multi-lingual embedding (see #658). However we support using Ollama/LMStudio with any embedder you can find on HF, which goes a long way.

You would indeed get better responses with the non-default. The default we have is super small so that is why we use it. Some other models like Jina can be more comprehensive but also are 300MB

@timothycarambat commented on GitHub (Feb 3, 2025): @kaanvur The default embedder is pretty bad at multi-lingual embedding (see #658). However we support using Ollama/LMStudio with _any_ embedder you can find on HF, which goes a long way. You would indeed get better responses with the non-default. The default we have is super small so that is why we use it. Some other models like Jina can be more comprehensive but also are 300MB
yindo commented 2026-02-22 18:27:30 -05:00
Author
Owner
Copy Link
Copy Source

@kaanvur commented on GitHub (Feb 3, 2025):

Thank you for all your answers and patience. I guess I need to try different combinations and see what the results are for the better results :) Thanks for your efforts!

@kaanvur commented on GitHub (Feb 3, 2025): Thank you for all your answers and patience. I guess I need to try different combinations and see what the results are for the better results :) Thanks for your efforts!
yindo changed title from [BUG]: In query mode, return answer citations with mach rates ranging from 0% to 3%. to [GH-ISSUE #3076] [BUG]: In query mode, return answer citations with mach rates ranging from 0% to 3%. 2026-06-05 14:43:46 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#1974
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?